Method, device and equipment for repairing theoretical line loss calculation abnormal data in power distribution network

Abstract

The invention relates to a method, a device and equipment for repairing theoretical line loss calculation abnormal data in a power distribution network. The method comprises the following steps: determining abnormal operation data and a normal operation data sequence adjacent to the abnormal operation data from acquired operation data in a power distribution network; performing sequence decomposition on the normal operation data sequence to obtain a plurality of operation data components; estimating the operation data of the operation moment corresponding to the abnormal operation data according to the plurality of operation data components to obtain estimated operation data; the estimated operation data is normal operation data at the operation moment corresponding to the estimated abnormal operation data; and calculating theoretical line loss data of the power distribution network at the operation moment corresponding to the abnormal operation data according to the estimated operation data. By adopting the method, the repair accuracy of the abnormal operation data can be improved, and then the accuracy of the determined theoretical line loss data in the power distribution network is improved.

Description

Technical Field

[0001] The present application relates to the technical field of distribution networks, and in particular to a method, device, and equipment for repairing abnormal data in theoretical line loss calculations in a distribution network. Background Art

[0002] With the development of information technology, a vast amount of measurement data has accumulated in distribution networks, including operational data such as voltage, current, power, and electricity consumption for distribution lines and distribution transformers. This data provides a sufficient basis for determining theoretical line loss data. However, due to metering device failures, communication interference, and other factors, measurement data is prone to loss and distortion, making it difficult to ensure the integrity and reliability of operational data, which in turn affects the reliability of theoretical line loss data.

[0003] Related technologies generally use methods such as mean filling and linear interpolation to repair abnormal operating data in distribution networks.

[0004] However, the related art has the problem of low accuracy in repairing abnormal operation data, which reduces the accuracy of the determined theoretical line loss data in the distribution network. Summary of the Invention

[0005] Based on this, the present application provides a method, device and equipment for repairing abnormal data of theoretical line loss calculation in a distribution network, which can improve the accuracy of repairing abnormal operation data, and thereby improve the accuracy of the determined theoretical line loss data in the distribution network.

[0006] In a first aspect, the present application provides a method for repairing abnormal data in theoretical line loss calculation in a distribution network, the method comprising:

[0007] Determining abnormal operating data and a normal operating data sequence adjacent to the abnormal operating data from operating data collected in the distribution network; the normal operating data sequence includes a first subnumber of normal operating data preceding the abnormal operating data and closest to the abnormal operating data, and / or a second subnumber of normal operating data following the abnormal operating data and closest to the abnormal operating data;

[0008] Decomposing the normal operation data sequence to obtain multiple operation data components;

[0009] estimating the operating data at the operating time corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data; the estimated operating data is the normal operating data at the operating time corresponding to the estimated abnormal operating data;

[0010] Based on the estimated operating data, the theoretical line loss data of the distribution network at the operating moment corresponding to the abnormal operating data is calculated.

[0011] In some embodiments, determining abnormal operating data and a sequence of normal operating data adjacent to the abnormal operating data from operating data collected in the power distribution network includes:

[0012] Sorting the collected operating data in ascending order to obtain sorted operating data;

[0013] The operation data with the target sequence number in the sorted operation data are determined as the low operation data threshold, and the operation data with the set sequence number are determined as the high operation data threshold; the set sequence number is the result of subtracting the target sequence number from the total number of collected operation data;

[0014] The operation data that is lower than a low operation data threshold or higher than a high operation data threshold among the collected operation data is determined as abnormal operation data, and a normal operation data sequence adjacent to the abnormal operation data is determined.

[0015] In some embodiments, the normal operation data sequence is decomposed to obtain multiple operation data components, including:

[0016] Decomposing the normal operating data sequence to obtain an operating data residual sequence and multiple operating data subsequences;

[0017] Determine the decomposition performance evaluation value of the normal operation data sequence based on the operation data residual sequence and the normal operation data sequence;

[0018] When the decomposition performance evaluation value is less than or equal to a preset evaluation threshold, determining the plurality of operating data subsequences into a plurality of operating data components;

[0019] When the decomposition performance evaluation value is greater than a preset evaluation threshold, the normal operation data sequence is re-decomposed to obtain a plurality of operation data sub-sequences, and the plurality of operation data sub-sequences are determined as a plurality of operation data components.

[0020] In some embodiments, the normal operation data sequence is subjected to sequence decomposition to obtain an operation data residual sequence and a plurality of operation data subsequences, including:

[0021] Obtaining a predetermined number of operation data components for estimating operation data at an operation time corresponding to the last abnormal operation data;

[0022] The normal operation data sequence is sequence-decomposed using a predetermined number of components to obtain an operation data residual sequence and multiple operation data subsequences.

[0023] In some embodiments, determining the decomposition performance evaluation value of the normal operating data sequence based on the operating data residual sequence and the normal operating data sequence includes:

[0024] Determining the ratios of the residual values ​​in the residual sequence of the operating data to the operating data in the normal operating data sequence as predetermined ratios;

[0025] The sum of the predetermined ratios is determined as a ratio accumulation value;

[0026] The ratio between the accumulated value of the ratio and the sum of the number of operating data in the normal operating data sequence is determined as the decomposition performance evaluation value of the normal operating data sequence.

[0027] In some embodiments, the normal operation data sequence is re-decomposed to obtain multiple operation data sub-sequences, including:

[0028] Constructing an initial optimal solution in the slime mold optimization algorithm according to a predetermined number of components for sequence decomposition of the normal operation data sequence, a preset quadratic penalty factor, and a convergence tolerance;

[0029] Determine the individual fitness in the slime mold optimization algorithm according to the maximum ratio and the minimum ratio of the ratios of each residual value in the running data residual sequence to each running data in the normal running data sequence;

[0030] Determine the random factor in the slime mold optimization algorithm based on the sine wave function;

[0031] According to the initial optimal solution, individual fitness and random factors, the target number of components is optimized;

[0032] The normal operation data sequence is decomposed using the target component number to obtain multiple operation data sub-sequences.

[0033] In some embodiments, estimating the operating data at the operating time corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data includes:

[0034] Determining each predicted operating data component according to each operating data component;

[0035] The sum of the predicted operating data components is determined as the estimated operating data at the operating time of the abnormal operating data.

[0036] In some embodiments, determining each predicted operating data component based on each operating data component includes:

[0037] Constructing a Lagrangian function based on the obtained optimization objective and constraints of the least squares support vector machine;

[0038] Determine the weight vector and bias in the constraint condition according to the normal operation data and the Lagrangian function in the operation data collected in the distribution network;

[0039] The predicted operating data components are determined based on the weight vector and bias in the constraint conditions and the operating data components.

[0040] In a second aspect, the present application provides a device for repairing abnormal data in theoretical line loss calculation in a distribution network, the device comprising:

[0041] an operating data determining module, configured to determine abnormal operating data and a sequence of normal operating data adjacent to the abnormal operating data from operating data collected from the distribution network; the sequence of normal operating data comprising a first sub-number of normal operating data preceding the abnormal operating data and closest to the abnormal operating data, and / or a second sub-number of normal operating data following the abnormal operating data and closest to the abnormal operating data;

[0042] A decomposition module, used for performing sequence decomposition on a normal operation data sequence to obtain multiple operation data components;

[0043] an estimating module, configured to estimate the operating data at the operating time corresponding to the abnormal operating data based on the plurality of operating data components to obtain estimated operating data; the estimated operating data being the normal operating data at the operating time corresponding to the estimated abnormal operating data;

[0044] The theoretical line loss data calculation module is used to calculate the theoretical line loss data of the distribution network at the operating time corresponding to the abnormal operating data based on the estimated operating data.

[0045] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the methods of the first aspect when executing the computer program.

[0046] In the technical solution provided in the embodiment of the present application, a plurality of operating data components are obtained by performing sequence decomposition on a normal operating data sequence, and the operating data at the operating moment corresponding to the abnormal operating data is estimated based on the plurality of operating data components to obtain estimated operating data. In this way, when determining the estimated operating data, the plurality of operating data components obtained by performing sequence decomposition on the normal operating data are used, which can avoid the influence of the unstable residual sequence in the normal operating data sequence on the determination of the estimated operating data. The estimated operating data is used as the repaired data of the abnormal operating data, which can improve the accuracy of the repair of the abnormal operating data, thereby improving the accuracy of the determined theoretical line loss data in the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 A schematic flow chart of a method for repairing abnormal data in theoretical line loss calculation in a distribution network provided by the first embodiment;

[0049] Figure 2 A flow chart of a method for repairing abnormal data in theoretical line loss calculation in a distribution network provided by the second embodiment;

[0050] Figure 3 A schematic flow chart of a method for repairing abnormal data in theoretical line loss calculation in a distribution network provided by the third embodiment;

[0051] Figure 4 A flowchart of a method for repairing abnormal data in theoretical line loss calculation in a distribution network provided by a fourth embodiment;

[0052] Figure 5 A flowchart of a repair process for abnormal operation data in a power distribution network provided in some embodiments;

[0053] Figure 6 A schematic diagram of the structure of a device for repairing abnormal data of theoretical line loss calculation in a distribution network provided by some embodiments;

[0054] Figure 7 A schematic structural diagram of a computer device provided in some embodiments. DETAILED DESCRIPTION

[0055] The following embodiments of the technical solution of the present application will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application and are therefore only examples and are not intended to limit the scope of protection of the present application.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0057] In the description of the embodiments of the present application, the technical terms "first" and "second" are only used to distinguish different objects and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "plurality" is more than two, unless otherwise clearly and specifically defined. In the description of the embodiments of the present application, "each" means each or each of a plurality, unless otherwise clearly and specifically defined.

[0058] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0059] In the description of the embodiments of this application, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0060] Currently, a significant amount of research is being conducted to improve the quality of source-side measurement data, aiming to enhance data integrity and reliability and better support the daily work of theoretical line loss professionals. Source-side data governance approaches fall into two main categories: statistical analysis, which utilizes mean filling and linear interpolation to repair abnormal operating data; and machine learning, which combines historical normal data with machine learning algorithms such as random forests, grayscale prediction, and long-short-term memory neural networks to build predictive models and output prediction results, using the predicted data to repair abnormal operating data.

[0061] However, while statistical analysis-based methods for repairing abnormal data are simple to apply, their results are inaccurate and fail to meet the demands of high-precision data repair. For example, mean filling can mask the volatility and dynamic characteristics of data, reducing its authenticity and information content. Linear interpolation methods are inaccurate for nonlinear or complex data fluctuations and may even introduce new biases. Machine learning-based methods for repairing abnormal data, however, have limited accuracy when used on more complex data.

[0062] Therefore, both the statistical analysis type abnormal operation data repair method and the machine learning type abnormal operation data repair method have the problem of low accuracy in repairing abnormal operation data, which leads to the use of the repaired data to calculate the theoretical line loss data, resulting in a reduction in the accuracy of the theoretical line loss data in the determined distribution network.

[0063] Figure 1 The flowchart of the method for repairing abnormal data of theoretical line loss calculation in the distribution network provided by the first embodiment is as follows: Figure 1 As shown, the method is applied to a computer device, and the method includes steps S101 to S104.

[0064] S101: Determine abnormal operation data and a normal operation data sequence adjacent to the abnormal operation data from operation data collected in a distribution network.

[0065] The collected operation data are arranged in the order of collection time.

[0066] One abnormal operation data corresponds to one normal operation data sequence, and one normal operation data sequence includes at least two normal operation data. The number of operation data in the normal operation data sequences corresponding to different abnormal operation data is the same. In some embodiments, the normal operation data sequence includes a first sub-number of normal operation data that is before the abnormal operation data and is closest to the abnormal operation data. In other embodiments, the normal operation data sequence includes a second sub-number of normal operation data that is after the abnormal operation data and is closest to the abnormal operation data. In still other embodiments, the normal operation data sequence includes a first sub-number of normal operation data that is before the abnormal operation data and is closest to the abnormal operation data, and a second sub-number of normal operation data that is after the abnormal operation data and is closest to the abnormal operation data. The first sub-number may be greater than or equal to the second sub-number. In still other embodiments, all normal operation data before an abnormal operation data in the operation data collected during a preset time period may be determined as the normal operation data sequence of the abnormal operation data.

[0067] A distribution network refers to a power grid that receives electrical energy from the transmission grid or regional power plants and distributes it locally or in a tiered manner according to voltage to various users through distribution facilities. For example, a distribution network may include medium and low voltage distribution networks. For example, a distribution network may include distribution lines (also known as distribution lines) and / or distribution transformers (also known as distribution transformers). Distribution lines are the lines that carry electricity from a step-down substation to a distribution transformer, or from a distribution substation to power users. A distribution transformer is a stationary electrical device within a distribution system that transforms AC voltage and current according to the law of electromagnetic induction to transmit AC power.

[0068] Exemplarily, the operating data may include at least one of the following: current data, voltage data, power data, and electricity data. The collected operating data may include all operating data collected within a preset time period. For example, if operating data is collected once every hour, the operating data collected within a year may include 365×24 operating data. For example, the preset time period may be a time period corresponding to a particular year. For another example, the preset time period may be the time period from the beginning of the year to the current time. For another example, the preset time period may be the time period from a preset time period before the current time to the current time.

[0069] Abnormal operation data may include at least one of a value of 0, a blank, data with a larger value, data with a smaller value, etc. Abnormal operation data is data in the collected operation data. Exemplarily, the collected operation data is one or more.

[0070] S102: Decompose the normal operation data sequence to obtain multiple operation data components.

[0071] Exemplarily, variational mode decomposition (VMD), empirical mode decomposition (EMD) or ensemble empirical mode decomposition (EEMD) can be used to perform sequence decomposition on the normal operation data sequence to obtain multiple operation data components.

[0072] For example, a predetermined number of components can be obtained and used to perform sequence decomposition on the normal operation data sequence to obtain multiple operation data components. In some embodiments, the predetermined number of components can be the number of operation data components used to estimate the operation data at the operation time corresponding to the last abnormal operation data. In other embodiments, the predetermined number of components can be a pre-set fixed number of components.

[0073] S103 . Estimate the operation data at the operation time corresponding to the abnormal operation data based on the multiple operation data components to obtain estimated operation data.

[0074] The estimated operation data is the normal operation data at the operation time corresponding to the estimated abnormal operation data.

[0075] In some embodiments, multiple estimated operating data components at the time of the abnormal operating data can be determined based on multiple operating data components; and estimated operating data can be determined based on the multiple estimated operating data components. Exemplarily, the multiple estimated operating data components can be fused (e.g., accumulated or weighted) to obtain the estimated operating data.

[0076] In other embodiments, fused operating data may be determined based on multiple operating data components. The operating data at the time corresponding to the abnormal operating data may be estimated based on the fused operating data to obtain estimated operating data. For example, the fused operating data may be obtained by fusion (e.g., accumulation or weighted fusion) of the multiple operating data components.

[0077] In some embodiments, the estimated operation data may be repair data of the abnormal operation data. In some embodiments, the estimated operation data may replace the abnormal operation data.

[0078] S104: Calculate theoretical line loss data of the distribution network at the operating moment corresponding to the abnormal operating data based on the estimated operating data.

[0079] Theoretical line loss refers to the energy loss incurred during the transmission, transformation, distribution, and sales phases of electricity from a power plant to a user. Exemplarily, theoretical line loss data may include the power lost (e.g., active power). Furthermore, theoretical line loss data may include the energy lost.

[0080] In some embodiments, the estimated operating data may be input into a trained theoretical line loss prediction model, and the theoretical line loss prediction model may be used to determine the theoretical line loss data of the distribution network at the operating moment of the abnormal operating data.

[0081] In other embodiments, the estimated operating data may be substituted into a theoretical line loss data calculation formula to obtain the theoretical line loss data of the distribution network at the time of operation of the abnormal operating data.

[0082] In the technical solution provided in the embodiment of the present application, a plurality of operating data components are obtained by performing sequence decomposition on a normal operating data sequence, and the operating data at the operating moment corresponding to the abnormal operating data is estimated based on the plurality of operating data components to obtain estimated operating data. In this way, when determining the estimated operating data, the plurality of operating data components obtained by performing sequence decomposition on the normal operating data are used, which can avoid the influence of the unstable residual sequence in the normal operating data sequence on the determination of the estimated operating data. The estimated operating data is used as the repaired data of the abnormal operating data, which can improve the accuracy of the repair of the abnormal operating data, thereby improving the accuracy of the determined theoretical line loss data in the distribution network.

[0083] Figure 2 The flowchart of the method for repairing abnormal data of theoretical line loss calculation in the distribution network provided by the second embodiment is as follows: Figure 2 As shown, the method is applied to a computer device, Figure 2 Example compared to Figure 1The difference between the embodiments is that S101 includes S1011 to S1013.

[0084] S1011 , sorting the collected operation data in ascending order to obtain sorted operation data.

[0085] For example, the collected operating data may be operating data within a preset time period. The collected operating data may be sorted in descending order or in descending order to obtain sorted operating data.

[0086] S1012: Determine the operating data with a sequence number equal to the target sequence number in the sorted operating data as a low operating data threshold, and determine the operating data with a sequence number equal to the set sequence number as a high operating data threshold.

[0087] The set sequence number is the result of subtracting the target sequence number from the total number of collected operating data.

[0088] Exemplarily, the target sequence number may be a fixed sequence number. Furthermore, illustratively, the target sequence number may be the result of multiplying the number of operating data included in a preset period by a preset ratio. For example, the preset ratio may be data that is greater than or equal to 0.001 and less than 0.05. For example, the preset ratio may be 0.001, 0.01, or 0.05, etc., which is not limited in the embodiments of the present application. In some embodiments, the preset ratio may be determined based on the detection accuracy of a detection device for the operating data.

[0089] Exemplarily, after the collected operating data are sorted from small to large, the operating data with a sequence number equal to the target sequence number is a low operating data threshold, and the operating data with a sequence number equal to the set sequence number is a high operating data threshold.

[0090] S1013: Determine, among the collected operating data, operating data that is lower than a low operating data threshold or higher than a high operating data threshold as abnormal operating data, and determine a normal operating data sequence adjacent to the abnormal operating data.

[0091] Abnormal operation data refers to abnormal data in theoretical line loss calculation. Abnormal operation data in theoretical line loss calculation refers to abnormal operation data used for theoretical line loss calculation.

[0092] In some embodiments, S1013 may include: obtaining the average value of the operating data other than the abnormal operating data in the collected operating data, and determining the operating data whose absolute value of the difference between the operating data below the low operating data threshold or above the high operating data threshold and the average value is greater than or equal to the set threshold as abnormal operating data.

[0093] Exemplarily, if an abnormal operating data is the first operating data among the collected operating data, the operating data collected before the collected operating data can be obtained, and multiple normal operating data before the abnormal operating data and closest to the abnormal operating data can be obtained from the previously collected operating data, so as to determine the normal operating data sequence adjacent to the abnormal operating data.

[0094] In the technical solution provided in the embodiment of the present application, the low operating data threshold and the high operating data threshold are flexibly determined based on the collected operating data, so that a preset proportion of abnormal operating data can be obtained from the collected operating data, avoiding the situation where the abnormal operating data cannot be effectively obtained due to unreasonable threshold setting. Therefore, the embodiment of the present application can improve the effectiveness of obtaining abnormal operating data.

[0095] Figure 3 The flowchart of the method for repairing abnormal data of theoretical line loss calculation in the distribution network provided by the third embodiment is as follows: Figure 3 As shown, the method is applied to a computer device, Figure 3 Example compared to Figure 1 The difference between the embodiments is that S102 includes S1021 to S1024.

[0096] S1021. Decompose the normal operation data sequence to obtain an operation data residual sequence and multiple operation data subsequences.

[0097] The dimensions of the normal operation data series, the dimensions of the operation data residual series, and the dimensions of multiple operation data subsequences are all the same.

[0098] In some embodiments, S1021 may include: performing sequence decomposition on the normal operation data sequence using a predetermined number of components to obtain multiple operation data subsequences; superimposing the respective operation data in the multiple operation data subsequences to obtain an operation data superposition sequence; and subtracting the operation data superposition sequence from the normal operation data sequence to obtain an operation data residual sequence.

[0099] In some embodiments, S1021 may include: obtaining a predetermined number of operation data components used to estimate the operation data at the operation time corresponding to the previous abnormal operation data; and performing sequence decomposition on the normal operation data sequence using the predetermined number of components to obtain an operation data residual sequence and multiple operation data subsequences. In this manner, since the detection devices that obtain the operation data are the same and the environmental parameters of the detection devices vary little, the number of operation data subsequences that can be decomposed is likely to be the same. Therefore, the predetermined number of operation data components used to estimate the operation data at the operation time corresponding to the previous abnormal operation data is determined as the number of components for the current sequence decomposition of the normal operation data sequence, thereby improving the decomposition efficiency of the current sequence decomposition of the normal operation data sequence.

[0100] In other embodiments, S1021 may include: if it is determined that the time difference between the moment of the current abnormal operation data and the moment of the previous abnormal operation data is less than or equal to a preset time difference, determining a predetermined number of operation data components used to estimate the operation data at the operation moment corresponding to the previous abnormal operation data as the number of components for the current sequence decomposition of the normal operation data sequence. If the time difference is greater than the preset time difference, determining a temperature difference between the moment of the current abnormal operation data and the moment of the previous abnormal operation data. If the temperature difference is less than or equal to a preset temperature, determining the predetermined number of operation data components used to estimate the operation data at the operation moment corresponding to the previous abnormal operation data as the number of components for the current sequence decomposition of the normal operation data sequence. If the temperature difference is greater than the preset temperature, determining the number of components corresponding to the moment of the current abnormal operation data where the temperature is less than or equal to the preset temperature before the moment of the current abnormal operation data as the number of components for the current sequence decomposition of the normal operation data sequence.

[0101] S1022. Determine a decomposition performance evaluation value of the normal operation data sequence based on the operation data residual sequence and the normal operation data sequence.

[0102] In some embodiments, S1022 may include: determining the ratios of each residual value in the operating data residual sequence to each operating data in the normal operating data sequence as predetermined ratios; determining the sum of each predetermined ratio as a ratio accumulation value; and determining the ratio between the ratio accumulation value and the sum of the number of operating data in the normal operating data sequence as a decomposition performance evaluation value of the normal operating data sequence.

[0103] In this way, each predetermined proportion can reflect the impact of each residual value on each operating data in the normal operating data sequence, and the ratio between the accumulated value of each predetermined proportion and the sum of the number of operating data in the normal operating data sequence is determined as the decomposition performance evaluation value. The decomposition performance evaluation value can take into account the impact of each residual value on each operating data, thereby improving the comprehensiveness of the residual evaluation and thereby improving the accuracy of the determined decomposition performance evaluation value.

[0104] In other embodiments, S1022 may include: determining the ratios of each residual value in the operating data residual sequence to each operating data in the normal operating data sequence as predetermined ratios, and determining the average value of each predetermined ratio as the decomposition performance evaluation value of the normal operating data sequence.

[0105] S1023. When the decomposition performance evaluation value is less than or equal to a preset evaluation threshold, determine the multiple operating data subsequences as multiple operating data components.

[0106] Exemplarily, the preset evaluation threshold may be a pre-set fixed threshold. In some embodiments, the preset evaluation threshold may be determined based on the evaluation accuracy of theoretical line loss data of the power distribution network.

[0107] S1024. When the decomposition performance evaluation value is greater than a preset evaluation threshold, re-decompose the normal operation data sequence to obtain multiple operation data sub-sequences, and determine the multiple operation data sub-sequences as multiple operation data components.

[0108] In some embodiments, the normal operation data sequence is re-decomposed to obtain multiple operation data sub-sequences, including: using a slime mold optimization algorithm (Slime Mould Algorithm, SMA) to optimize the predetermined number of components to obtain a target number of components, and using the target number of components to decompose the normal operation data sequence to obtain multiple operation data sub-sequences.

[0109] In some embodiments, a slime mold optimization algorithm is used to optimize the predetermined number of components to obtain a target number of components, including: constructing an initial optimal solution in the slime mold optimization algorithm based on the predetermined number of components used for sequence decomposition of the normal operating data sequence, a preset quadratic penalty factor, and a convergence tolerance; determining the individual fitness in the slime mold optimization algorithm based on the maximum ratio and the minimum ratio of the ratios of each residual value in the operating data residual sequence to each operating data in the normal operating data sequence; determining a random factor in the slime mold optimization algorithm based on a sine wave function; and optimizing the target number of components based on the initial optimal solution, the individual fitness, and the random factor.

[0110] In this way, the slime mold optimization algorithm simulates the intelligent behavior of slime molds during the foraging process, including the exploration and exploitation phases. It can strike a balance between global search and local optimization, thereby avoiding falling into local optimal solutions. By optimizing the predetermined number of components through the slime mold optimization algorithm, it can efficiently find the optimal target number of components, and then dynamically adjust the number of components according to data characteristics to ensure the rationality of the decomposition results.

[0111] Figure 4 The flowchart of the method for repairing abnormal data of theoretical line loss calculation in the distribution network provided by the fourth embodiment is as follows: Figure 4 As shown, the method is applied to a computer device, Figure 4 Example compared to Figure 1 The difference between the embodiments is that S103 includes S1031 to S1032.

[0112] S1031. Determine each predicted operating data component according to each operating data component.

[0113] In some embodiments, a least squares support vector machine (LSSVM) may be used to predict each operating data component to obtain each predicted operating data component. In other embodiments, a long short-term memory (LSTM) network or a bidirectional long short-term memory network may be used to predict each operating data component to obtain each predicted operating data component.

[0114] The following example uses the least squares support vector machine to predict each operating data component to illustrate the implementation method of S1031: construct a Lagrangian function based on the obtained optimization objectives and constraints of the least squares support vector machine; determine the weight vector and bias in the constraints based on the normal operating data and the Lagrangian function in the operating data collected in the distribution network; determine each predicted operating data component based on the weight vector and bias in the constraints and each operating data component.

[0115] In this way, by constructing the Lagrangian function, the optimization problem of the least squares support vector machine is transformed into a mathematical problem with constraints. The introduction of the Lagrangian function enables the optimization objective and constraints to be solved under a unified framework, simplifying the complexity of the problem.

[0116] S1032: Determine the sum of the predicted operating data components as the estimated operating data at the operating time of the abnormal operating data.

[0117] In the technical solution provided in the embodiment of the present application, by predicting each operating data component, each predicted operating data component obtained can represent the characteristics of each component, reducing the situation where effective features are omitted, and thereby improving the accuracy of the determined estimated operating data.

[0118] The embodiment of the present application proposes a new method for repairing abnormal data in theoretical line loss calculation of distribution network by combining variational mode decomposition (VMD), slime mold optimization algorithm (SMA), and least squares support vector machine (LSSVM), aiming to improve the quality and reliability of source data and better support the implementation of theoretical line loss calculation and loss reduction simulation analysis work in theoretical line loss professional.

[0119] The embodiment of the present application proposes a novel method for repairing abnormal data in the theoretical line loss calculation of a distribution network. Taking the current data of a single distribution line as an example, the overall flow chart is as follows, including seven parts: data collection, abnormal data location, variational mode decomposition (VMD), residual sequence calculation, threshold analysis, LSSVM subsequence prediction, and abnormal data repair. The repair of abnormal measurement data related to the theoretical line loss calculation of conductors and distribution transformers in the distribution network, including voltage, current, active power, reactive power, active electricity, reactive electricity, etc., can be carried out according to this process.

[0120] Figure 5 A flowchart of a repair process for abnormal operation data in a distribution network provided in some embodiments is shown in FIG. Figure 5 As shown, in the repair process of abnormal operation data, data collection is first performed to obtain the collected operation data; abnormal operation data positioning is performed to obtain abnormal operation data; a normal operation data sequence adjacent to the abnormal operation data is obtained, and variational mode decomposition (VMD) is used to decompose the normal operation data sequence to obtain multiple operation data subsequences. For example, the multiple operation data subsequences are subsequence 1, subsequence 2, subsequence 3, and up to subsequence K, etc., where K is an integer greater than or equal to 2.

[0121] After obtaining a plurality of operating data subsequences, residual sequence calculation is performed to obtain an operating data residual sequence; and a decomposition performance evaluation value of the normal operating data sequence is determined based on the operating data residual sequence and the normal operating data sequence.

[0122] A threshold analysis is performed on the decomposition performance evaluation value. If the decomposition performance evaluation value does not exceed the threshold (i.e., the preset evaluation threshold), a least squares support vector machine (LSSVM) is used (on multiple operating data subsequences) to perform subsequence prediction, obtaining multiple predicted operating data components. If the decomposition performance evaluation value exceeds the threshold (i.e., the preset evaluation threshold), the slime mold optimization algorithm (SMA) parameter optimization is used to obtain the target number of components. The normal operating data sequence is then decomposed using the target number of components to obtain multiple operating data subsequences. Subsequence prediction is then performed using LSSVM (on multiple operating data subsequences) to obtain multiple predicted operating data components.

[0123] After obtaining multiple predicted operation data components, the sum of the predicted operation data components is determined as the estimated operation data at the operation moment of the abnormal operation data; the estimated operation data is used to repair the abnormal operation data, for example, the abnormal operation data is replaced by the estimated operation data.

[0124] The following uses current data as an example to illustrate how to implement the above steps:

[0125] In data collection, the source system is combined to collect the historical 24-hour current data of the target distribution line for one year. , where the subscript i represents the number of days, and the value range is , j represents the time, and its value range is .

[0126] In locating abnormal operation data, first, Sort by value to form a set , from the above data collection, we can know that the set There are 8760 data in total; secondly, the positioning set The 876th and 7884th data are respectively Finally, the judgment logic is designed to locate abnormal operation data, as follows: . Iterate over the collection Repeat the judgment logic in the above formula for all data in the traversal. After the traversal is completed, all current data with a value of 0 are Treat it as abnormal operation data and sort it in chronological order to form a collection .

[0127] In the abnormal operation data repair, the repair steps of the first abnormal operation data in the set are taken as an example. The repair of other abnormal operation data in the set can be carried out according to the following steps.

[0128] First, VMD decomposition is performed. For example, the time dimension is collected All the normal current data before forming a collection , as follows: .

[0129] (1) VMD decomposition

[0130] 1) Collect by time dimension All the normal current data before forming a collection , illustratively, It should be noted that here Taking the above-mentioned normal current data as an example to form a normal operating data sequence, in other embodiments, the normal operating data sequence may be determined by other methods listed above.

[0131] 2) Gather All data in are regarded as a continuous time signal. Assuming that it can be decomposed into K subsequences by VMD (including the primary subsequence, the K subsequences are the multiple running data subsequences mentioned above), then the expression of the kth subsequence is: ; The value range of t is 1 to The number of all normal current data before, that is, the number of data in a subsequence and the number of sets The number of current data remains consistent. can be Replacement; Phase is a non-decreasing function, and ; represents the envelope function (envelope amplitude); represents the kth subsequence.

[0132] 3) The bandwidth of each IMF component (i.e., each subsequence) It can be estimated according to the Casson criterion, as follows: ;in, represents the maximum deviation of the instantaneous frequency from the center, Indicates the maximum instantaneous frequency deviation, Indicates the highest frequency of the envelope signal.

[0133] 4) Under the constraint that the sum of the components is equal to the input signal, the sum of the estimated bandwidths of the components is minimized. After a series of transformations, the following constrained variational model is constructed:

[0134] ;

[0135] in, is the set of subsequences; is the set of corresponding center frequencies; is the partial derivative of the function with respect to time t; is the pulse function.

[0136] (2) Residual sequence calculation

[0137] Residual sequence The proportion of the original time series is an important indicator for measuring the performance of VMD decomposition. The smaller the proportion, the better the VMD decomposition effect. The specific calculation formula is as follows: ;in, express All the normal current data before (i.e. one way to realize the normal operating data sequence above), The number of current data in all the previous normal current data is N.

[0138] (3) Threshold analysis

[0139] The threshold index F (i.e., the above-mentioned decomposition performance evaluation value) is designed to quantitatively evaluate the decomposition performance of VMD, as follows: ; Among them, the value range of F is 0 to 1. The smaller the value of F, the better the decomposition performance of VMD, and the larger the value of F, the worse the decomposition performance of VMD. Represents the tth data in the residual sequence; express The tth data among all the previous normal current data.

[0140] (4) Threshold judgment

[0141] Analyze whether the value of F meets the requirements of VMD decomposition effectiveness. If the decomposition performance evaluation value F is less than or equal to the preset evaluation threshold, it meets the requirements of VMD decomposition effectiveness. If so, jump directly to step (6), otherwise jump to step (5).

[0142] (5) SMA parameter optimization

[0143] The calculation formula of the decomposition performance evaluation value F is defined as the objective function. Combined with the slime mold optimization algorithm, the key parameters of VMD decomposition are set, such as the number of modes K and the quadratic penalty factor And convergence tolerance (Tolerance, Tol), etc.

[0144] 1) The position update formula in the slime mold optimization algorithm is: ;in, Indicates the position of the current optimal individual (i.e., the optimal solution in the slime mold optimization algorithm), which is the key parameters decomposed by VMD, such as the number of modes K, the quadratic penalty factor , a multidimensional column vector composed of convergence tolerance Tol, etc.; is the weight; and are two randomly selected solutions to introduce randomness; Represents the random factor, which is used to adjust the size of random perturbations; Represents a random number to further increase randomness; represents the worst solution. For example, the quadratic penalty factor You can refer to similar signal decomposition scenarios and make preliminary settings Tol can be set in advance according to the calculation accuracy requirements of the algorithm.

[0145] 2) The adaptive weight function is determined as follows ;in, It represents the fitness value of the i-th individual in the population. It is calculated by substituting the VMD parameters corresponding to the individual (such as the number of modes K, the quadratic penalty factor α, etc.) into the objective function (i.e., the threshold index F). It is used to measure the quality of the VMD decomposition effect under the individual parameter combination. : Indicates the fitness value of the best individual in the current population, which is also calculated based on the objective function F and represents the decomposition effect corresponding to the best parameter combination currently searched.

[0146] Adjust the search step size by the following logic: If the individual fitness is poor Big value, increase, make Calculating The weight of the individual is increased, the movement range of the individual in the search space is increased, and the search range is expanded; if the individual fitness is poor Small value, Reduce, the individual movement range is reduced, and focus on local fine search. The impact on position update ultimately achieves dynamic adjustment of search step length, balancing the algorithm's global exploration and local development capabilities. Dynamically adjust the search step size, individuals with poor fitness expand the search range, and high-quality individuals focus on local development.

[0147] 3) You can use an oscillation contraction strategy, such as introducing a sine wave function to periodically adjust the search direction. For example, you can use a random factor To periodically adjust the search direction, ; The period T is used to control the switching frequency between global search and local development, and are the amplitude and vertical offset in the sine wave function, respectively.

[0148] (6) LSSVM subsequence prediction

[0149] The first subsequence after VMD decomposition For example, combined with subsequence data, the LSSVM regression model output The values ​​are as follows:

[0150] 1) The constraints of the LSSVM model are ;in, is the weight vector, is the bias term; is the first subsequence at time t, is the first subsequence at time t+1; for and predicted results. A nonlinear mapping function that maps an input vector to a high-dimensional feature space. For example, the kernel function is used to achieve implicit mapping without explicitly calculating high-dimensional space operations.

[0151] By mapping the input vector to a high-dimensional feature space, it is possible to process nonlinear relationships and enhance the model's expressiveness. In processing nonlinear relationships, in the original low-dimensional space, the subsequence data It may show nonlinear distribution characteristics, which makes it difficult to achieve effective fitting using linear models. By mapping to high-dimensional feature space, the nonlinear problem in the original space can be transformed into a linear problem in high-dimensional space, so that LSSVM can complete nonlinear prediction tasks with the help of linear models. In enhancing the expressive power of the model, high-dimensional feature space can provide richer feature representation forms and effectively expand the model hypothesis space, which enables LSSVM to learn more complex patterns in subsequence data and improve the accuracy of the model. The prediction accuracy can be improved, thereby restoring the original signal characteristics more accurately.

[0152] 2) The optimization goal of the LSSVM model is The optimization goal of LSSVM is to find the weight vector that minimizes the model prediction error. and the bias term b, where is the error term, which represents the deviation between the model prediction value and the true value.

[0153] In order to solve the optimization problem, Lagrange multipliers are introduced , define the Lagrangian function as:

[0154] ;in, is a regularization parameter used to control the complexity of the model.

[0155] 3) Solving model parameters

[0156] Lagrangian function Taking the partial derivative and setting it equal to zero, we get the following system of equations:

[0157] ;

[0158] By solving the above equations, the model parameters can be obtained and , thus determining the LSSVM model.

[0159] By solving this system of equations, we can not only obtain and , and can also be calculated based on the equations Calculate the weight vector , while combining Solving for residuals .for and Since this set of equations is an analytical solution obtained by taking the partial derivative of the Lagrangian function, it is a direct optimization solution and does not require additional iterative optimization updates. This process has completed the parameter determination based on the optimization goal of LSSVM (such as minimizing structural risk), so we get and After that, the LSSVM model can be determined directly without further iteration.

[0160] 4) Prediction

[0161] The predicted value is expressed as follows: .

[0162] For other subsequences Repeat the above steps 1)-4) to get The predicted value of .

[0163] (6) Repair of abnormal operation data

[0164] The prediction value of each subsequence is summed to complete The repairs are as follows: .

[0165] This embodiment of the application provides a novel method for repairing abnormal operating data in distribution network theoretical line loss calculations. Leveraging the exceptional data decomposition capabilities of VMD, the complexity of the original time series is reduced, enabling the LSSVM prediction model to produce better predictions. Designing a threshold metric to quantitatively evaluate VMD decomposition performance and optimizing VMD parameters using the SMA optimization algorithm can further improve VMD decomposition performance.

[0166] The abnormal operation data repair method proposed in the embodiment of the present application only analyzes one-dimensional data. Compared with the machine learning model, it does not require feature engineering and the model construction is simpler.

[0167] The abnormal operation data repair method proposed in the embodiment of the present application introduces VMD decomposition and SMA optimization algorithm. Compared with the statistical analysis type abnormal operation data repair method and the single machine learning type data repair method, the repair result is more accurate.

[0168] Based on the same inventive concept, embodiments of the present application also provide a device for repairing abnormal data in the theoretical line loss calculation of a distribution network, which is used to implement the aforementioned method for repairing abnormal data in the theoretical line loss calculation of a distribution network. The implementation solution provided by this device is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the device for repairing abnormal data in the theoretical line loss calculation of a distribution network provided below can be found in the above-mentioned limitations of the method for repairing abnormal data in the theoretical line loss calculation of a distribution network, and will not be repeated here.

[0169] In an exemplary embodiment, Figure 6 A schematic diagram of a device for repairing abnormal data of theoretical line loss calculation in a distribution network provided in some embodiments, such as Figure 6 As shown, the device 600 for repairing abnormal data of theoretical line loss calculation in the distribution network includes:

[0170] An operating data determining module 601 is configured to determine abnormal operating data and a sequence of normal operating data adjacent to the abnormal operating data from operating data collected from the distribution network; the sequence of normal operating data includes a first sub-number of normal operating data preceding the abnormal operating data and closest to the abnormal operating data, and / or a second sub-number of normal operating data following the abnormal operating data and closest to the abnormal operating data;

[0171] Decomposition module 602, configured to perform sequence decomposition on the normal operation data sequence to obtain multiple operation data components;

[0172] An estimation module 603 is configured to estimate the operating data at the operating time corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data; the estimated operating data is the normal operating data at the operating time corresponding to the estimated abnormal operating data;

[0173] The theoretical line loss data calculation module 604 is used to calculate the theoretical line loss data of the distribution network at the operating time corresponding to the abnormal operating data based on the estimated operating data.

[0174] In some embodiments, the operation data determination module 601 includes a sorting unit, a threshold determination unit and an operation data determination unit; the sorting unit is used to sort the collected operation data in ascending order to obtain sorted operation data; the threshold determination unit is used to determine the operation data with a sequence number of the target sequence number in the sorted operation data as a low operation data threshold, and determine the operation data with a sequence number of the set sequence number as a high operation data threshold; the set sequence number is the result of subtracting the target sequence number from the total number of collected operation data; the set sequence number is the result of subtracting the target sequence number from the total number of collected operation data; the operation data determination unit is used to determine the operation data in the collected operation data that is lower than the low operation data threshold or higher than the high operation data threshold as abnormal operation data, and determine the normal operation data sequence adjacent to the abnormal operation data.

[0175] In some embodiments, the decomposition module 602 includes a decomposition unit, an evaluation value determination unit, and an operating data component determination unit; the decomposition unit is used to perform sequence decomposition on the normal operating data sequence to obtain an operating data residual sequence and multiple operating data subsequences; the evaluation value determination unit is used to determine the decomposition performance evaluation value of the normal operating data sequence based on the operating data residual sequence and the normal operating data sequence; the operating data component determination unit is used to determine the multiple operating data subsequences as multiple operating data components when the decomposition performance evaluation value is less than or equal to a preset evaluation threshold; when the decomposition performance evaluation value is greater than the preset evaluation threshold, the normal operating data sequence is re-decomposed to obtain multiple operating data subsequences, and the multiple operating data subsequences are determined as multiple operating data components.

[0176] In some embodiments, the decomposition unit is also used to obtain a predetermined number of operating data components used to estimate the operating data at the operating moment corresponding to the previous abnormal operating data; and use the predetermined number of components to perform sequence decomposition on the normal operating data sequence to obtain an operating data residual sequence and multiple operating data subsequences.

[0177] In some embodiments, the evaluation value determination unit is also used to determine the ratios of each residual value in the operating data residual sequence to each operating data in the normal operating data sequence as predetermined ratios; determine the sum of each predetermined ratio as the ratio cumulative value; and determine the ratio between the ratio cumulative value and the sum of the number of operating data in the normal operating data sequence as the decomposition performance evaluation value of the normal operating data sequence.

[0178] In some embodiments, the operating data component determination unit is further used to construct an initial optimal solution in the slime mold optimization algorithm based on a predetermined number of components used for sequence decomposition of the normal operating data sequence, a preset quadratic penalty factor, and a convergence tolerance; determine the individual fitness in the slime mold optimization algorithm based on the maximum ratio and the minimum ratio of the ratios of each residual value in the operating data residual sequence to each operating data in the normal operating data sequence; determine the random factor in the slime mold optimization algorithm based on a sine wave function; optimize and obtain a target number of components based on the initial optimal solution, the individual fitness, and the random factor; and use the target number of components to sequence decompose the normal operating data sequence to obtain multiple operating data sub-series.

[0179] In some embodiments, the estimation module 603 includes a predicted operating data component determination unit and an estimated operating data determination unit; wherein the predicted operating data component determination unit is used to determine each predicted operating data component based on each operating data component; the estimated operating data determination unit is used to determine the sum of each predicted operating data component as the estimated operating data at the operating moment of the abnormal operating data.

[0180] In some embodiments, the predicted operating data component determination unit is also used to construct a Lagrangian function based on the optimization objectives and constraints of the acquired least squares support vector machine; determine the weight vector and bias in the constraints based on the normal operating data and the Lagrangian function in the operating data collected in the distribution network; determine each predicted operating data component based on the weight vector and bias in the constraints, as well as each operating data component.

[0181] The description of the above device embodiment is similar to the description of the above method embodiment and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiment of this application, please refer to the description of the method embodiment of this application for understanding.

[0182] Each module in the aforementioned device for repairing abnormal data in theoretical line loss calculations in a distribution network can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0183] In an exemplary embodiment, Figure 7This is a schematic diagram of the structure of a computer device provided in some embodiments. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. The wireless communication method can be implemented via wireless fidelity (Wi-Fi), a mobile cellular network, near-field communication (NFC), or other technologies. When executed by the processor, the computer program implements a method for repairing abnormal data in theoretical line loss calculations in a distribution network. The display unit of the computer device is used to produce a visual image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0184] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0185] For example, a computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method of any of the above embodiments when executing the computer program.

[0186] For example, in an exemplary embodiment, the processor is used to execute a computer program to implement: determining abnormal operating data and a normal operating data sequence adjacent to the abnormal operating data from the operating data collected in the distribution network; the normal operating data sequence includes a first sub-number of normal operating data that is before the abnormal operating data and is closest to the abnormal operating data, and / or a second sub-number of normal operating data that is after the abnormal operating data and is closest to the abnormal operating data; performing sequence decomposition on the normal operating data sequence to obtain multiple operating data components; estimating the operating data at the operating moment corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data; the estimated operating data is the normal operating data at the operating moment corresponding to the estimated abnormal operating data; and calculating the theoretical line loss data of the distribution network at the operating moment corresponding to the abnormal operating data based on the estimated operating data.

[0187] In one embodiment, a computer-readable storage medium is provided, and when a computer program is executed by a processor, the computer program implements the steps of the method provided in any of the above embodiments.

[0188] For example, in an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: determining abnormal operating data and a normal operating data sequence adjacent to the abnormal operating data from the operating data collected in the distribution network; the normal operating data sequence includes a first sub-number of normal operating data before the abnormal operating data and closest to the abnormal operating data, and / or a second sub-number of normal operating data after the abnormal operating data and closest to the abnormal operating data; performing sequence decomposition on the normal operating data sequence to obtain multiple operating data components; estimating the operating data at the operating moment corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data; the estimated operating data is the normal operating data at the operating moment corresponding to the estimated abnormal operating data; and calculating the theoretical line loss data of the distribution network at the operating moment corresponding to the abnormal operating data based on the estimated operating data.

[0189] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps of the method provided in any of the above embodiments are implemented.

[0190] For example, in an exemplary embodiment, a computer program product is provided, including a computer program, which implements the following steps when executed by a processor: determining abnormal operating data and a normal operating data sequence adjacent to the abnormal operating data from the operating data collected in the distribution network; performing sequence decomposition on the normal operating data sequence to obtain multiple operating data components; estimating the operating data at the operating moment corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data; and calculating the theoretical line loss data of the distribution network at the operating moment of the abnormal operating data based on the estimated operating data.

[0191] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0192] The processor, each functional module or each functional unit in any embodiment of the present application may include any one or more of the following integrations: a general-purpose processor, an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a central processing unit (CPU), a graphics processing unit (GPU), an embedded neural network processing unit (NPU), a controller, a microcontroller, a microprocessor, a programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, a data processing logic based on quantum computing, an artificial intelligence (AI) processor, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0193] The memory or computer-readable storage medium in any embodiment of the present application may include at least one of a non-volatile memory and a volatile memory. Non-volatile memory includes the integration of one or more of the following: Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Ferromagnetic Random Access Memory (FRAM), Flash Memory, Magnetic Surface Storage, Optical Disc, Compact Disc Read-Only Memory (CD-ROM), Magnetic Tape, Floppy Disk, Flash Memory, Optical Storage, High-density Embedded Non-volatile Memory, Resistive Random Access Memory (ReRAM), Magnetoresistive Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), Graphene Memory, Volatile Memory, etc. Volatile memory includes one or more of the following: random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can come in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0194] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0195] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for repairing abnormal data in theoretical line loss calculation in a distribution network, characterized in that: The method comprises: determining abnormal operating data and a normal operating data sequence adjacent to the abnormal operating data from operating data collected from the power distribution network; the normal operating data sequence including a first subnumber of normal operating data preceding the abnormal operating data and closest to the abnormal operating data, and / or a second subnumber of normal operating data following the abnormal operating data and closest to the abnormal operating data; Decomposing the normal operation data sequence to obtain multiple operation data components; estimating the operating data at the operating time corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data; the estimated operating data is the estimated normal operating data at the operating time corresponding to the abnormal operating data; The theoretical line loss data of the distribution network at the operating moment corresponding to the abnormal operating data is calculated based on the estimated operating data.

2. The method according to claim 1, characterized in that The determining of abnormal operation data and a normal operation data sequence adjacent to the abnormal operation data from the operation data collected in the distribution network includes: Sorting the collected operating data in ascending order to obtain sorted operating data; Determining the operating data having a sequence number equal to a target sequence number in the sorted operating data as a low operating data threshold, and determining the operating data having a sequence number equal to a set sequence number as a high operating data threshold; the set sequence number is the result of subtracting the target sequence number from the total number of the collected operating data; The operation data that is lower than the low operation data threshold or higher than the high operation data threshold in the collected operation data is determined as the abnormal operation data, and a normal operation data sequence adjacent to the abnormal operation data is determined.

3. The method according to claim 1 or 2, characterized in that Decomposing the normal operation data sequence to obtain multiple operation data components includes: Decomposing the normal operating data sequence to obtain an operating data residual sequence and a plurality of operating data subsequences; determining a decomposition performance evaluation value of the normal operating data sequence according to the operation data residual sequence and the normal operating data sequence; In a case where the decomposition performance evaluation value is less than or equal to a preset evaluation threshold, determining the multiple operating data subsequences as the multiple operating data components; When the decomposition performance evaluation value is greater than the preset evaluation threshold, the normal operation data sequence is re-decomposed to obtain a plurality of operation data sub-sequences, and the plurality of operation data sub-sequences are determined as the plurality of operation data components.

4. The method according to claim 3, characterized in that The performing sequence decomposition on the normal operation data sequence to obtain an operation data residual sequence and a plurality of operation data subsequences includes: Obtaining a predetermined number of operation data components for estimating operation data at an operation time corresponding to the last abnormal operation data; The normal operation data sequence is sequence decomposed using the predetermined number of components to obtain the operation data residual sequence and the multiple operation data subsequences.

5. The method according to claim 3, characterized in that The determining, based on the operating data residual sequence and the normal operating data sequence, a decomposition performance evaluation value of the normal operating data sequence includes: determining the ratios of the residual values ​​in the residual sequence of the operating data to the operating data in the normal operating data sequence as predetermined ratios; Determining the sum of the predetermined ratios as a ratio accumulation value; The ratio between the accumulated value of the ratios and the sum of the number of operating data in the normal operating data sequence is determined as the decomposition performance evaluation value of the normal operating data sequence.

6. The method according to claim 3, characterized in that The re-decomposition of the normal operation data sequence to obtain multiple operation data sub-sequences includes: constructing an initial optimal solution in the slime mold optimization algorithm according to a predetermined number of components for sequence decomposition of the normal operation data sequence, a preset quadratic penalty factor, and a convergence tolerance; Determining the individual fitness in the slime mold optimization algorithm according to the maximum ratio and the minimum ratio of the ratios of each residual value in the operating data residual sequence to each operating data in the normal operating data sequence; Determining a random factor in the slime mold optimization algorithm according to a sine wave function; Optimizing the target component quantity according to the initial optimal solution, the individual fitness and the random factor; The normal operation data sequence is sequence-decomposed using the target number of components to obtain a plurality of operation data subsequences.

7. The method according to claim 1 or 2, characterized in that The estimating the operation data at the operation time corresponding to the abnormal operation data based on the multiple operation data components to obtain estimated operation data includes: Determining each predicted operating data component based on each of the operating data components; The sum of the predicted operating data components is determined as the estimated operating data at the operating moment of the abnormal operating data.

8. The method according to claim 7, characterized in that Determining each predicted operating data component according to each operating data component includes: Constructing a Lagrangian function based on the obtained optimization objective and constraints of the least squares support vector machine; Determining a weight vector and a bias in the constraint condition according to normal operating data in the operating data collected in the distribution network and the Lagrangian function; The predicted operating data components are determined according to the weight vector and the bias in the constraint conditions and the operating data components.

9. A device for repairing abnormal data of theoretical line loss calculation in a distribution network, characterized in that: The device comprises: an operating data determining module, configured to determine abnormal operating data and a sequence of normal operating data adjacent to the abnormal operating data from operating data collected in the distribution network; the sequence of normal operating data comprising a first sub-number of normal operating data preceding the abnormal operating data and closest to the abnormal operating data, and / or a second sub-number of normal operating data following the abnormal operating data and closest to the abnormal operating data; a decomposition module, configured to perform sequence decomposition on the normal operating data sequence to obtain a plurality of operating data components; an estimating module, configured to estimate the operating data at the operating time corresponding to the abnormal operating data based on the multiple operating data components to obtain estimated operating data; the estimated operating data is the estimated normal operating data at the operating time corresponding to the abnormal operating data; The theoretical line loss data calculation module is used to calculate the theoretical line loss data of the distribution network at the operating time corresponding to the abnormal operating data based on the estimated operating data.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images